On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,051 papers · 148 categories

Trend · papers per month

count % of corpus

48 results for “Unseen Instances”

Heterogeneous Graph-based Knowledge Transfer for Generalized Zero-shot Learningcs.LG

HGKT transfers knowledge from seen to unseen classes in GZSL without prior unseen class info.

problem Learning to classify unseen classes in GZSL.

method Structured heterogeneous graph with graph neural network for knowledge transfer.

result Achieves state-of-the-art results on public benchmark datasets.

zero-shot learning knowledge transfer graph neural network Wasserstein barycenter heterogeneous graph

2019-11-20abs ↗pdf ↗

Instance based Generalization in Reinforcement Learningcs.LG

RL agents fail to generalize to unseen environments, even when dynamics are similar.

problem RL agents fail to generalize to unseen environments despite similar dynamics.

method Analyzed policy learning in POMDPs, formalized training dynamics as instances, and introduced a shared belief representation over an ensemble of specialized policies.

result Maximizing rewards induces instance-specific policies that are suboptimal on the training set.

Reinforcement Learning Generalization Policy Learning Markov Decision Processes Shared Belief Representation

2020-11-02abs ↗pdf ↗

The CM Algorithm for the Maximum Mutual Information Classifications of Unseen Instancescs.LG

CM algorithm improves MMI classifications for unseen instances.

problem Improving classification accuracy for unseen instances using MMI criterion.

method Introduces CM algorithm for MMI classifications, combining semantic and Shannon channels for matching.

result Achieves high mutual information (99%) with minimal iterations in low-dimensional feature spaces.

Maximum Mutual Information Channels Matching Unseen Instances Semantic Information Low-Dimensional Feature Spaces

2019-01-28abs ↗pdf ↗

How to Explain Individual Classification Decisionsstat.ML

After building a classifier with modern tools of machine learning we typically have a black box at hand that is able to predict well for unseen data. Thus, we get an answer to the question what is the most likely label of a given unseen data point. However, most methods will provide no answer why the model predicted th…

2009-12-06abs ↗pdf ↗

Deep Weakly-supervised Anomaly Detectioncs.LG

PReNet detects seen and unseen anomalies using pairwise relations.

problem Detecting unseen anomalies in semi-supervised learning.

method Pairwise Relation prediction Network (PReNet) learns anomaly and normal patterns.

result PReNet significantly outperforms nine competing methods in anomaly detection.

anomaly detection semi-supervised learning weak supervision pairwise relations unseen anomalies

2019-10-30abs ↗pdf ↗

Zero-Shot Learning via Semantic Similarity Embeddingcs.CV

In this paper we consider a version of the zero-shot learning problem where seen class source and target domain data are provided. The goal during test-time is to accurately predict the class label of an unseen target domain instance based on revealed source domain side information (\eg attributes) for unseen classes. …

2015-09-15abs ↗pdf ↗

TGG: Transferable Graph Generation for Zero-shot and Few-shot Learningcs.LG

TGG improves zero-shot and few-shot learning by explicitly modeling and utilizing seen-unseen domain relations.

problem Lack of data in unseen domains hinders generalization in zero-shot and few-shot learning.

method TGG generates explicit instance-level graphs to model and utilize seen-unseen domain relations, addressing domain shift.

result TGG outperforms existing methods in zero-shot, generalized zero-shot, and few-shot learning.

Transferable Graph Generation Zero-shot Learning Few-shot Learning Domain Shift Relation Modeling

2019-08-30abs ↗pdf ↗

Weakly Supervised One-Shot Detection with Attention Similarity Networksstat.ML

Neural network models that are not conditioned on class identities were shown to facilitate knowledge transfer between classes and to be well-suited for one-shot learning tasks. Following this motivation, we further explore and establish such models and present a novel neural network architecture for the task of weakly…

2018-01-10abs ↗pdf ↗

Learning to Generalize to Unseen Tasks with Bilevel Optimizationcs.LG

A new framework improves few-shot classification by learning to generalize to unseen classes.

problem Training metric-based meta-learning approaches for few-shot classification often fails to generalize to unseen classes.

method Proposes a bilevel optimization framework to explicitly constrain meta-training to reduce unseen classification error.

result Significantly improves performance on unseen classes compared to episodic training.

meta-learning few-shot classification bilevel optimization unsupervised learning generalization

2019-08-05abs ↗pdf ↗

Kernel Mean Embedding of Instance-wise Predictions in Multiple Instance Regressionstat.ML

Proposes a new method for MIR using kernel mean embeddings.

problem Multiple instance regression (MIR) where bags contain multiple instances with a single label.

method Computes kernel mean embeddings of predicted label distributions and learns a regressor from these embeddings.

result Better results than baseline instance-MIR across all datasets, state-of-the-art on two.

multiple instance regression kernel mean embedding instance-MIR distribution regression

2019-04-24abs ↗pdf ↗

Few-shot Learning for Unsupervised Feature Selectioncs.LG

Proposes a few-shot learning method for feature selection without labeled data.

problem Feature selection in unlabeled data with limited instances.

method Uses Concrete random variables and permutation-invariant neural networks to select features from multiple source tasks.

result Outperforms existing methods in feature selection performance.

few-shot learning unsupervised feature selection Concrete random variables permutation-invariant neural networks

2021-07-02abs ↗pdf ↗

Learning from Aggregate Observationsstat.ML

The paper extends multiple instance learning to multiclass and regression problems.

problem Learning from aggregate observations where supervision is given to sets of instances.

method Probabilistic framework for various aggregate observations, including classification and regression.

result The proposed estimator has nice convergence properties under mild assumptions.

multiple instance learning aggregate observations multiclass classification regression maximum likelihood

2020-04-14abs ↗pdf ↗

Computing Human-Understandable Strategiescs.GT

Algorithms for equilibrium computation generally make no attempt to ensure that the computed strategies are understandable by humans. For instance the strategies for the strongest poker agents are represented as massive binary files. In many situations, we would like to compute strategies that can actually be implement…

2016-12-19abs ↗pdf ↗

Feature-Critic Networks for Heterogeneous Domain Generalizationcs.LG

New method learns domain-invariant features for unseen domains.

problem Learning models that generalize to unseen domains with different statistics.

method Learning to learn approach, training a domain-invariant feature extractor.

result Method outperforms state-of-the-art solutions in both domain generalization and heterogeneous domain generalization.

domain generalization feature extraction learning to learn heterogeneous domains

2019-01-31abs ↗pdf ↗

Adaptive Adjustment with Semantic Feature Space for Zero-Shot Recognitioncs.CV

Adaptive adjustment of semantic feature space improves zero-shot recognition.

problem Domain shift and hubness problems in zero-shot recognition.

method Adaptive adjustment of semantic feature space.

result Remarkable performance improvement compared to existing methods.

zero-shot recognition semantic feature space domain shift hubness adaptive adjustment

2019-03-30abs ↗pdf ↗

Learning the Compositional Spaces for Generalized Zero-shot Learningcs.LG

This paper studies the problem of Generalized Zero-shot Learning (G-ZSL), whose goal is to classify instances belonging to both seen and unseen classes at the test time. We propose a novel space decomposition method to solve G-ZSL. Some previous models with space decomposition operations only calibrate the confident pr…

2018-10-17abs ↗pdf ↗

OwMatch: Conditional Self-Labeling with Consistency for Open-World Semi-Supervised Learningcs.LG

OwMatch improves open-world semi-supervised learning by self-labeling and consistency.

problem Misclassification of unseen classes in open-world semi-supervised learning.

method Conditional self-labeling and open-world hierarchical thresholding.

result OwMatch enhances performance across known and unknown classes.

semi-supervised learning open-world self-labeling consistency unseen classes

2024-11-04abs ↗pdf ↗

Generalization on the Unseen, Logic Reasoning and Degree Curriculumcs.LG

The paper explores how different network architectures learn logical functions under GOTU, finding that a min-degree-interpolator is learned.

problem Learning logical functions with a focus on generalization on the unseen.

method Study of different network architectures trained by SGD under GOTU.

result For sparse functions and certain network models, a min-degree-interpolator is learned on the unseen.

Logical functions Generalization on the unseen Min-degree-interpolator Network architectures SGD

2023-01-30abs ↗pdf ↗

Learning with Sets in Multiple Instance Regression Applied to Remote Sensingstat.ML

A new method for MIR in remote sensing without assuming a prime instance per bag.

problem Multiple Instance Regression in remote sensing with high variability.

method Treats each bag as a set of instances and learns to map each bag to its unique label using all instances.

result Outperforms previous state-of-the-art on three real-world datasets.

Multiple Instance Regression Remote Sensing Set-based Learning Attention Mechanism Order-invariant Operation

2019-03-18abs ↗pdf ↗

Evolving Multi-label Classification Rules by Exploiting High-order Label Correlationcs.LG

This paper improves multi-label classification by leveraging high-order label correlations.

problem Improving accuracy in multi-label classification tasks using label correlations.

method Exploiting high-order label correlations through a supervised learning classifier system (UCS) and label powerset (LP) strategy.

result The proposed method outperforms other LP-based methods on multiple benchmark datasets.

multi-label classification label correlation supervised learning UCS label powerset

2020-07-22abs ↗pdf ↗

Leveraging Pre-Trained 3D Object Detection Models For Fast Ground Truth Generationcs.LG

Training 3D object detectors for autonomous driving has been limited to small datasets due to the effort required to generate annotations. Reducing both task complexity and the amount of task switching done by annotators is key to reducing the effort and time required to generate 3D bounding box annotations. This paper…

2018-07-16abs ↗pdf ↗

Continual Rare-Class Recognition with Emerging Novel Subclassescs.LG

RaRecognize learns to recognize rare classes in a stream of data.

problem Learning to recognize rare classes in a continuous stream of data.

method Estimates a general decision boundary, learns individual rare subclasses, flags new subclasses.

result RaRecognize outperforms state-of-the-art baselines on real-world datasets.

Rare class recognition Continuous learning Emerging subclasses Decision boundary Minority class

2019-06-28abs ↗pdf ↗

Data-driven Projection Generation for Efficiently Solving Heterogeneous Quadratic Programming Problemsstat.ML

Efficiently solves heterogeneous QPs by reducing variables using instance-specific projections.

problem Solving high-dimensional quadratic programming problems efficiently.

method Data-driven framework with a graph neural network generating projections tailored to each QP instance.

result Produces high-quality solutions with reduced computation time, outperforming existing methods.

Quadratic Programming Heterogeneous Problems Projection Graph Neural Networks Efficient Solutions

2025-10-30abs ↗pdf ↗

Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognitioncs.LG

Unified approach to continual learning using generative replay and open set recognition.

problem Catastrophic interference and recognition of out-of-distribution data in deep neural networks.

method Probabilistic approach based on variational inference in a deep autoencoder model, using generative replay and open set recognition.

result The approach significantly alleviates catastrophic interference and distinguishes out-of-distribution data.

continual learning catastrophic interference generative replay open set recognition variational inference

2019-05-28abs ↗pdf ↗

Algorithm Runtime Prediction: Methods & Evaluationcs.AI

Perhaps surprisingly, it is possible to predict how long an algorithm will take to run on a previously unseen input, using machine learning techniques to build a model of the algorithm's runtime as a function of problem-specific instance features. Such models have important applications to algorithm analysis, portfolio…

2012-11-05abs ↗pdf ↗

Differentiable Cutting-plane Layers for Mixed-integer Linear Optimizationmath.OC

Differentiable cutting-plane layers solve parametric mixed-integer linear optimization problems.

problem Solving parametric mixed-integer linear optimization problems with changing data.

method Introducing cutting-plane layers (CPLs) for differentiable cutting-plane generation.

result The algorithm computes solutions with low integrality gaps and generalizes to unseen instances.

Mixed-integer linear optimization Differentiable cutting-plane Parametric optimization Cut generation

2023-11-06abs ↗pdf ↗

Learning to Dispatch for Job Shop Scheduling via Deep Reinforcement Learningcs.LG

Deep RL learns effective job shop scheduling rules from raw features.

problem Designing effective priority dispatching rules for job shop scheduling is challenging.

method End-to-end deep reinforcement learning using Graph Neural Networks.

result Agent learns high-quality dispatching rules from raw features and generalizes well to unseen instances.

Job Shop Scheduling Deep Reinforcement Learning Graph Neural Networks Priority Dispatching Rules

2020-10-23abs ↗pdf ↗

Meta-learning One-class Classifiers with Eigenvalue Solvers for Supervised Anomaly Detectionstat.ML

Meta-learning improves anomaly detection with few labeled instances.

problem High requirement of training data for neural network-based anomaly detection.

method Meta-learning framework with one-class classification and generalized eigenvalue problem.

result Meta-learning method achieves better performance than existing methods on various datasets.

meta-learning anomaly detection few-shot learning one-class classification neural networks

2021-03-01abs ↗pdf ↗

Meta Dropout: Learning to Perturb Features for Generalizationcs.LG

Meta-dropout learns to perturb training examples for better generalization.

problem Achieving good generalization in unseen test examples.

method Meta-learning a noise generator to perturb latent features of training examples.

result Meta-dropout significantly improves generalization performance on few-shot classification datasets.

meta-learning generalization noise generation perturbation few-shot learning

2019-05-30abs ↗pdf ↗

InstaNAS: Instance-aware Neural Architecture Searchcs.LG

Conventional Neural Architecture Search (NAS) aims at finding a single architecture that achieves the best performance, which usually optimizes task related learning objectives such as accuracy. However, a single architecture may not be representative enough for the whole dataset with high diversity and variety. Intuit…

2018-11-26abs ↗pdf ↗

Modified Meta-Thompson Sampling for Linear Bandits and Its Bayes Regret Analysisstat.ML

Modified Meta-TS for linear contextual bandits reduces regret.

problem Optimizing decision-making in dynamic environments with context vectors.

method Meta-TSLB algorithm for linear contextual bandits, analyzing Bayes regret.

result Derives an

O((m+\log(m))\sqrt{n\log(n)})

bound on Bayes regret.

Meta-Thompson Sampling Linear Bandits Contextual Bandits Bayes Regret Meta-Learning

2024-09-10abs ↗pdf ↗

Semantic Channel and Shannon's Channel Mutually Match for Multi-Label Classificationcs.LG

A group of transition probability functions form a Shannon's channel whereas a group of truth functions form a semantic channel. Label learning is to let semantic channels match Shannon's channels and label selection is to let Shannon's channels match semantic channels. The Channel Matching (CM) algorithm is provided f…

2018-05-02abs ↗pdf ↗

Exploiting Generalization in Offline Reinforcement Learning via Unseen State Augmentationscs.LG

New method finds unseen states for RL, improving performance.

problem Offline RL struggles with unseen states and actions.

method Value-informed state perturbations and filtering.

result Improved performance in offline RL tasks.

Offline RL Unseen states Value-informed perturbations Epistemic uncertainty Conservative Q-values

2023-08-07abs ↗pdf ↗

Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Taskscs.LG

Bayesian TAML balances meta-knowledge and task-specific learning for imbalanced tasks.

problem Meta-learning approaches struggle with varying instance and class numbers and distributional differences.

method Bayesian inference framework with variational inference to adaptively balance meta-knowledge and task-specific learning.

result Bayesian TAML significantly outperforms existing meta-learning approaches on imbalanced datasets.

meta-learning Bayesian inference variational inference imbalanced tasks out-of-distribution

2019-05-30abs ↗pdf ↗

Parameterizing Branch-and-Bound Search Trees to Learn Branching Policiescs.LG

New method learns better branching policies for MILP problems.

problem Improving branch and bound search for solving MILP problems.

method Imitates strong branching rule with parameterized state of B&B search tree.

result Generalized policies outperform current state-of-the-art.

MILP Branch and Bound Imitation Learning Branching Policies Parameterization

2020-02-12abs ↗pdf ↗

Reinforcement-Learning-Based Variational Quantum Circuits Optimization for Combinatorial Problemscs.LG

RL optimizes quantum circuit parameters for combinatorial problems.

problem Optimizing quantum circuit parameters for combinatorial problems.

method Reinforcement Learning (RL) to train a policy network.

result RL policy reduces optimality gap by up to 8.61.

Reinforcement Learning Quantum Computing Variational Quantum Circuits Combinatorial Optimization Quantum Approximate Optimization Algorithm (QAOA)

2019-11-11abs ↗pdf ↗

Learning to Optimize Variational Quantum Circuits to Solve Combinatorial Problemscs.LG

Paper proposes machine learning to optimize QAOA for combinatorial problems.

problem Optimizing QAOA parameters for solving combinatorial optimization problems.

method Develops two machine learning approaches: RL and KDE to learn optimal QAOA parameters.

result Reduces optimality gap by up to 30.15 compared to other optimizers.

Quantum Computing QAOA Reinforcement Learning Kernel Density Estimation Combinatorial Optimization

2019-11-25abs ↗pdf ↗

Generalizing Across Domains via Cross-Gradient Trainingcs.LG

We present CROSSGRAD, a method to use multi-domain training data to learn a classifier that generalizes to new domains. CROSSGRAD does not need an adaptation phase via labeled or unlabeled data, or domain features in the new domain. Most existing domain adaptation methods attempt to erase domain signals using technique…

2018-04-28abs ↗pdf ↗

DUET: Optimizing Training Data Mixtures via Feedback from Unseen Evaluation Taskscs.LG

DUET optimizes LLM training data mixtures using feedback from unseen tasks.

problem Optimizing LLM performance on unseen tasks without prior data knowledge.

method Interleaves influence function with Bayesian optimization for data selection.

result DUET converges to optimal training data mixture for unseen tasks.

LLM training data feedback Bayesian optimization influence function

2025-02-01abs ↗pdf ↗

Multiple-Kernel Dictionary Learning for Reconstruction and Clustering of Unseen Multivariate Time-seriescs.LG

MKD learns MTS attributes to reconstruct and cluster unseen classes.

problem Reconstructing and clustering unseen multivariate time-series.

method Multiple-kernel dictionary learning (MKD) to learn semantic attributes.

result MKD provides interpretable reconstruction and high clustering performance.

multiple-kernel dictionary learning multivariate time-series reconstruction clustering unseen classes

2019-03-05abs ↗pdf ↗

Zero-shot Learning via Simultaneous Generating and Learningcs.LG

A deep generative model learns from seen and unseen classes without explicit training data.

problem Overcoming zero-shot learning with unseen classes.

method Variational auto-encoder with class-specific multi-modal prior, iteratively generating and learning unseen data.

result Outperforms models trained only on seen classes and state-of-the-art methods.

zero-shot learning generative model variational auto-encoder unsupervised learning missing data

2019-10-21abs ↗pdf ↗

Offline Meta Learning of Explorationcs.LG

Meta-agent learns effective exploration from offline data.

problem Design a meta-agent to quickly maximize reward in unseen tasks.

method Bayesian RL approach with adaptive neural belief estimate.

result Meta-agent learns effective exploration behavior from diverse tasks.

Offline Meta Reinforcement Learning Bayesian Reinforcement Learning Exploration Strategy Meta-Agent MDP Ambiguity

2020-08-06abs ↗pdf ↗

Group-based Learning of Disentangled Representations with Generalizability for Novel Contentscs.LG

Sensory data are often comprised of independent content and transformation factors. For example, face images may have shapes as content and poses as transformation. To infer separately these factors from given data, various ``disentangling'' models have been proposed. However, many of these are supervised or semi-super…

2018-09-07abs ↗pdf ↗

Towards Neural Network Patching: Evaluating Engagement-Layers and Patch-Architecturescs.LG

The study evaluates neural network patching techniques to adapt models to concept drift.

problem Adapting neural network models to handle concept drift in nonstationary environments.

method Investigated different engagement layers and patch architectures to enhance neural network patching.

result Identified generally applicable heuristics for parametrizing the patching procedure.

neural network patching concept drift nonstationary environments engagement layers patch architectures

2018-12-09abs ↗pdf ↗

Instance-Wise Monotonic Calibration by Constrained Transformationcs.LG

New method calibrates neural network predictions for better reliability.

problem Improper probability estimates from deep networks leading to unreliable predictions.

method Proposes a constrained optimization approach for a monotonic calibration map.

result Achieves state-of-the-art performance across various datasets and models.

probability calibration deep learning monotonicity post-hoc calibration neural networks

2025-07-09abs ↗pdf ↗

Message-Aware Graph Attention Networks for Large-Scale Multi-Robot Path Planningcs.RO

A new network learns to prioritize messages for efficient multi-robot path planning.

problem Efficient path planning and coordination for large-scale multi-robot systems.

method Message-Aware Graph Attention Network (MAGAT) incorporating attention mechanisms.

result MAGAT achieves performance close to a coupled centralized expert algorithm.

Graph Neural Networks Multi-Robot Systems Path Planning Attention Mechanisms Decentralized Coordination

2020-11-26abs ↗pdf ↗

DART: Dropouts meet Multiple Additive Regression Treescs.LG

Multiple Additive Regression Trees (MART), an ensemble model of boosted regression trees, is known to deliver high prediction accuracy for diverse tasks, and it is widely used in practice. However, it suffers an issue which we call over-specialization, wherein trees added at later iterations tend to impact the predicti…

2015-05-07abs ↗pdf ↗

OpenKI: Integrating Open Information Extraction and Knowledge Bases with Relation Inferencecs.IR

Proposes OpenKI for better web-scale knowledge extraction and alignment.

problem Combining OpenIE and KB for web-scale knowledge extraction and alignment.

method Instance-level inference using neighborhood information from KB and OpenIE extractions, with attention mechanisms.

result Significantly improves performance on OpenIE extractions and semi-structured data.

knowledge extraction knowledge bases open information extraction relation inference entity-level inference

2019-04-12abs ↗pdf ↗